Abstract: A hub is disclosed that includes a network, first and second modules, a processor, a failsafe bus, and an alarm wire bus. The network interface component communicates data with a medical device. The first module interfaces with a first communications channel, and the second module interfaces with a second communications channel. The processor packages the data from the medical device into at least one packet. The processor communicates with one of the first and second modules to communicate one or more packets over one of the first and second communications channels. The failsafe bus signals from the hub to the medical device when a fatal-error fault condition of the hub has occurred. The alarm wire bus signals from the hub to the medical device when an alarm condition of the hub has occurred to cause the medical device to execute at least one mitigation.

Abstract: Detection models are distributed as needed to a select subset of image capture devices, based on geographic location and time of the image capture devices. Each image capture device which receives a model processes image data according to the model, and provides a notification of any detected instances of an event the model is trained to detect. Distribution and detection can be based on historical data or live data.

Abstract: Certain aspects of the present disclosure provide techniques for identifying a minimal, or at least reduced, set of representative calibration paths in radio frequency (RF) circuits and calibrating other calibration paths based on calibration codes used for the representative calibration paths. An example method generally includes receiving a calibration data set including measurements associated with each calibration path of a plurality of calibration paths in an RF circuit. Based on a clustering model and the calibration data set, a plurality of calibration clusters is generated. From each respective calibration cluster of the plurality of calibration clusters, a respective representative calibration path for is selected for the respective calibration cluster. Generally, calibration codes generated for the representative calibration path are applicable to other calibration paths in the calibration cluster.

Abstract: Detection models are distributed as needed to a select subset of image capture devices, based on geographic location and time of the image capture devices. Each image capture device which receives a model processes image data according to the model, and provides a notification of any detected instances of an event the model is trained to detect. Distribution and detection can be based on historical data or live data.

Abstract: Systems and methods by a telematics server are provided. The method includes receiving, over a network, training data including model input data and a known output label corresponding to the model input data from a first device, training a centralized machine-learning model using the training data, determining, by the centralized machine-learning model, an output label prediction certainty based on the model input data, determining an increase in the output label prediction certainty over a prior predicted output label certainty of the centralized machine-learning model, and sending, over the network, a machine-learning model update to a second device in response to determining that the increase in the output label prediction certainty is greater than an output label prediction increase threshold.

Abstract: Methods for vehicle data collection by image analysis are provided. An example method involves positioning a camera in a vehicle to be pointed toward a field of interest in the vehicle, capturing an image of the field of interest with the camera, identifying a region of interest in the image that is expected to convey vehicle information, and running an image processing model over the region of interest to extract vehicle information from the image.

Abstract: Systems and methods for training image processing models for vehicle data collection by image analysis are provided. An example method involves accessing an image of a field of interest in a vehicle captured by a camera in the vehicle, providing a user interface to, display the image, receive input that defines a region of interest in the image that is expected to convey vehicle information, and receive input that assigns a label to the region of interest that associates the region of interest with an image processing model that is to be trained to extract a type of vehicle information from the region of interest, and contributing the image, labelled with the region of interest and the label associating the region of interest to the image processing model, to a training data library to train the image processing model.

Abstract: Systems and methods by a telematics server are provided. The method includes receiving, over a network, training data including model input data and a known output label corresponding to the model input data from a first device, training a centralized machine-learning model using the training data, determining, by the centralized machine-learning model, an output label prediction certainty based on the model input data, determining an increase in the output label prediction certainty over a prior predicted output label certainty of the centralized machine-learning model, and sending, over the network, a machine-learning model update to a second device in response to determining that the increase in the output label prediction certainty is greater than an output label prediction increase threshold.

Abstract: An input/output (I/O) expansion adapter and a method by an I/O expansion adapter are provided. The method includes receiving raw I/O expansion data from an I/O expander coupled thereto, processing the raw I/O expansion into processed I/O expansion data, and sending the processed I/O expansion data to the telematics device.

Abstract: Detection models are distributed as needed to a select subset of image capture devices, based on geographic location and time of the image capture devices. Each image capture device which receives a model processes image data according to the model, and provides a notification of any detected instances of an event the model is trained to detect. Distribution and detection can be based on historical data or live data.

Abstract: Detection models are distributed as needed to a select subset of image capture devices, based on geographic location and time of the image capture devices. Each image capture device which receives a model processes image data according to the model, and provides a notification of any detected instances of an event the model is trained to detect. Distribution and detection can be based on historical data or live data.

Abstract: Methods for vehicle data collection by image analysis are provided. An example method involves positioning a camera in a vehicle to be pointed toward a field of interest in the vehicle, capturing an image of the field of interest with the camera, identifying a region of interest in the image that is expected to convey vehicle information, and running an image processing model over the region of interest to extract vehicle information from the image.

Abstract: Systems and methods for training image processing models for vehicle data collection by image analysis are provided. An example method involves accessing an image of a field of interest in a vehicle captured by a camera in the vehicle, providing a user interface to, display the image, receive input that defines a region of interest in the image that is expected to convey vehicle information, and receive input that assigns a label to the region of interest that associates the region of interest with an image processing model that is to be trained to extract a type of vehicle information from the region of interest, and contributing the image, labelled with the region of interest and the label associating the region of interest to the image processing model, to a training data library to train the image processing model.

Abstract: Systems and methods for vehicle data collection by image analysis are provided. An example method involves positioning a camera in a vehicle to be pointed toward a field of interest in the vehicle, capturing an image of the field of interest with the camera, identifying a region of interest in the image that is expected to convey vehicle information, and running an image processing model over the region of interest to extract vehicle information from the image.

Abstract: Systems are disclosed for a communication system optimized for low data volume communications. In embodiments of the invention, a terminal in the communication system is configured to send and a receiver in the network infrastructure is configured to receive bursts containing pilot waveforms that include time spacing and/or phase changes between short sequences. The time spacing and/or phase changes can be selected to reduce out-of-phase autocorrelation and achieve robust performance.

Abstract: Systems are disclosed for a communication system optimized for low data volume communications. In embodiments of the invention, a terminal in the communication system is configured to send and a receiver in the network infrastructure is configured to receive bursts containing pilot waveforms that include time spacing and/or phase changes between short sequences. The time spacing and/or phase changes can be selected to reduce out-of-phase autocorrelation and achieve robust performance.

Abstract: One feature pertains to a digitally controlled oscillator (DCO) that comprises a variable capacitor and noise reduction circuitry. The variable capacitor has a variable capacitance value that controls an output frequency of the DCO. The variable capacitance value is based on a first bank capacitance value provided by a first capacitor bank, a second bank capacitance value provided by a second capacitor bank, and an auxiliary bank capacitance value provided by an auxiliary capacitor bank. The noise reduction circuitry is adapted to adjust the variable capacitance value by adjusting the auxiliary bank capacitance value while maintaining at least one of the first bank capacitance value and/or the second bank capacitance value substantially unchanged. Prior to adjusting the variable capacitance value, the noise reduction circuitry may determine that a received input DCO control word transitions across a capacitor bank sensitive boundary.

Abstract: One feature pertains to a digitally controlled oscillator (DCO) that comprises a variable capacitor and noise reduction circuitry. The variable capacitor has a variable capacitance value that controls an output frequency of the DCO. The variable capacitance value is based on a first bank capacitance value provided by a first capacitor bank, a second bank capacitance value provided by a second capacitor bank, and an auxiliary bank capacitance value provided by an auxiliary capacitor bank. The noise reduction circuitry is adapted to adjust the variable capacitance value by adjusting the auxiliary bank capacitance value while maintaining at least one of the first bank capacitance value and/or the second bank capacitance value substantially unchanged. Prior to adjusting the variable capacitance value, the noise reduction circuitry may determine that a received input DCO control word transitions across a capacitor bank sensitive boundary.

Abstract: A phase locked loop (PLL) device is configurable in an analog phase locked loop and a hybrid analog-digital phase locked loop. In an analog mode, at least a phase detector, an analog loop filter, and a voltage controlled oscillator (VCO), are connected to form an analog loop. In a digital mode, at least the phase detector, the voltage controlled oscillator (VCO), a time to digital converter (TDC), a digital loop filter and a digital to analog converter (DAC) are connected to form the hybrid digital-analog loop.

Abstract: Techniques for performing analog calibration of a receiver to optimize a second-order input intercept point (IIP2). In an aspect, a signal generator modeling an interferer is coupled to an adjustable input of a receiver, e.g., a gate bias voltage of a mixer. For example, the signal generator output may be a single-tone on-off keying (OOK) modulated signal. The mixer mixes the signal down to baseband, wherein an analog correlator correlates the down-converted signal with the known sequence of bits used to perform the OOK modulation. The analog correlation output is then provided to drive the bias voltage in the mixer, e.g., one or more gate voltages of transistors in the differential mixer, to optimize the overall receiver IIP2. Further aspects of the disclosure provide for calibrating receivers having multiple LNA's, and also dual or diversity receivers having multiple receive paths.